Abstract Controversial results concerning the association between a polymorphism rs6822844 in the interleukin (IL) 21 (IL-21) gene and rheumatoid arthritis (RA) have existed. A meta-analysis to confirm above relationships is necessary to be performed immediately. We conducted a search in the PubMed database, covering all papers published up to 20 October 2018. Overall, six case–control studies with 3244 cases and 3431 healthy controls were included. Odds ratios (ORs) and 95% confidence intervals (CIs) were used to assess the strength of this association. Publication bias was assessed with both Egger’s and Begg’s tests. After calculation, we found that IL-21 rs6822844 polymorphism could decrease RA risk in overall genetic models (allelic contrast: OR = 0.77, 95% CI = 0.62–0.97, P=0.024; TG versus GG: OR = 0.68, 95% CI = 0.50–0.92, P=0.013, and dominant genetic model: OR = 0.72, 95% CI = 0.55–0.94, P=0.016). Similarly, stratified analysis by race, source of control, significantly decreased association was found in Asians, Caucasians and hospital-based (HB) control source. Finally, in the subgroup analysis of rheumatoid factor (RF) and anti-citrullinated protein antibody (ACPA) status, poorly decreased relationship was detected between IL-21 rs6822844 polymorphism and RF negative and ACPA positive RA risk, respectively. No obvious evidence of publication bias was detected in overall analysis. In summary, our study indicated that IL-21 rs6822844 polymorphism was significantly associated with decreased RA susceptibility.